This will be controversial.

(Bridger) #121

Meh, this might be true. I guess it depends upon what you mean when you say “best.” But this is not my argument.

Some people have said that K/D is a useless statistic. That is demonstrably false.

(Ragoo) #122

So much text, sorry I can’t possibly read it all.

My opinion:

Don’t show K/D on casual servers. Allow us to enable showing K/D on our custom servers tho. Sort by XP (cause that’s probably most important statistic for showing your value for the team) and then also show K/D as it’s the most interesting ,easy to grasp and common statistic beside XP.

Other statistics are interesting as well like accuracy (with different weapons like Quake Live) and class specific XP and stuff like amount of revives, restored HP etc etc but this is obviously information overload if you have it in the game and only useful for spectator mode.

(JeP) #123

Bridger > that’s my point : they’re different way to evaluate a player as the “best”.

(tokamak) #124

[QUOTE=JeP;273310]Don’t you understand that neither XP nor K/D will tell you who is the best player ?
[/QUOTE]

XP comes incredibly close while K/D doesn’t even approximate the player’s skill.

You didn’t demonstrate anything, you bailed. I suggest you read that post you ignored again, if you then find you may need any further elaboration on why you’re completely wrong then I’m all here.

(Apples) #125

All this XP stuff is very balance dependant tho, while K/D is not… If they dont reward mucho for a kill, then I can easily have more XP than a good killer pushing in the front and helping his mates while doing side objectives (construct a mg nest / blow a secondary door etc) which are IMO a waste of time if you arent yet at this part of the map (ie : the frontline is far from this obj).

K/D is not at all a good stat to show that a player helped his team (camper can have a high K/D by simply not dying much and only kill like 4 peoples during the whole game… pointless), there we agree and I never said the opposite, but I dont think you proved that XP is somehow a good statistics neither, as its all based on strong hypothesis that this XP will be representative of the teamplay, which neither you nor me know at this point. I’m not playing with words here I’m enumerating simple facts.

The XP - XP/min can be very biased and is not really a strong index of how helpfull you were for your team, I’ve seen this many times in ETQW, when the highest XP was just farming somewhere, and his team still loose (and his team has less average XP than the winning side), so no, XP is not that “good”. I can also argue that in order to kill many enemies, you have to be on the frontline, or at least between their spawn and the objective, which makes a big killer even more valuable has he usually play side by side with you, while you can gain xp far from the front, by farming deployable and stuffs.

The counter arguments are easy when all is based on speculation, I’m not very familiar with game design, meta-games and such litterate words, I just describe facts and reality based on experience in gaming, and I might be wrong, but you wont convince me till you show me some facts (and guess what… you cant as the game isnt out yet, I’m so smart!) about XP being a valuable index (I didnt say more valuable than K/D, just valuable in a general definition), and about a team of poor killers winning against a team of better killer, etc etc.

Peace

(Bridger) #126

[QUOTE=tokamak;273345]XP comes incredibly close while K/D doesn’t even approximate the player’s skill.

You didn’t demonstrate anything, you bailed. I suggest you read that post you ignored again, if you then find you may need any further elaboration on why you’re completely wrong then I’m all here.[/QUOTE]

If high K/D correlates with team wins, then it’s not a meaningless statistic. That’s the only way to prove it one way or the other. Time will tell.

(Apples) #127

You can barely correlate a ratio VS a constant, or at least you cant take conclusions from this, this is basic rules in statistics as you will never know if the high K/D came from a high Kill or a Low death, thats what Tok is spinning about I guess. On the other hand you can maybe correlate a High kill count with wins, and there i’ll be more in agreement with you.

Peace

(tokamak) #128

You will never know if the high K/D came from a high Kill or a Low death

Aye, and that is providing killing is an indicative skillset, which it isn’t.

This correlation is much weaker than xp/min, xp/min also has a much stronger causal link. There are countless instances where a high K/D doesn’t win you the match while there are very few instances where high team xp doesn’t win you the map. Xp is has a very strong causal link to winning the match, if you’re defending well the xp keeps pouring in, if you complete main objectives you receive a lot, anything you do that rewards xp has a high probability of helping towards a win.

K/D however does not have this probability, people might have sacrificed themselves to defend or protect an objective (which would translate into a higher xp but in a lower K/D). There are very few instances where a player is faced with either winning the match or foregoing that and receiving more xp.

K/D only favours one particular playstyle and one particular skill set, and neither are necessarily contribute to a more functional team. It’s an absolutely useless indicator.

(Bridger) #129

[QUOTE=Apples;273360]You can barely correlate a ratio VS a constant, or at least you cant take conclusions from this, this is basic rules in statistics as you will never know if the high K/D came from a high Kill or a Low death, thats what Tok is spinning about I guess. On the other hand you can maybe correlate a High kill count with wins, and there i’ll be more in agreement with you.

Peace[/QUOTE]

This is true, but I speculate that a character who consistently has a high K/D ratio will also have more wins than characters with lower K/D ratios. This is very messy in the short term due to many other factors, but over a few hundred games i think it would become clear.

Both High kills AND low deaths help you win the game. Killing the enemy prevents them from killing you. Staying alive lets you complete objectives.

As was said before: you Can’t complete the objective if you are dead. Someone who is very good at killing and staying alive is going to be more useful to a team than someone who isn’t.*

*yes the guy who hides all game and gets no kills is the exception, but I’d guess it’s a very rare one.

(Senethro) #130

K/D can be manipulated easily through behaviours that don’t win games. Kills/min is harder to do so and is more useful than k/d anyway.

(Bridger) #131

True, Kills/min is probably a better indicator than K/D.

(Herandar) #132

Why not be leet and make it kills/rounds fired?

(tokamak) #133

Because that would only reward snipers, shotguns and kill-stealers.

(Herandar) #134

Kills/Headshots then. I appear to have forgotten the [/sarcasm] tag above. Hope it was obvious there…

(tokamak) #135

If this forum had a higher average IQ then the sarcasm would be more apparent but right now it’s just drowned in the noise of inane shouting.

(Ragoo) #136

rofl, you are really the most arrogant dick on this forum.

(Nail) #137

not even close, there’s way more arrogant dicks around here
top ten maybe, but not #1

:tongue:

(tokamak) #138

Yeah it’s a real shame the level of this forum gives me so much reason to be an arrogant dick.

(DarkangelUK) #139

You’re not arrogant, just pig headed and can’t see past yourself… big difference. You don’t actually have anything to be arrogant about.

(tokamak) #140

Oh I’m willing to concede points if the arguments are convincing, they just rarely are.